By Andrew Coleman*
The introduction and expansion of government pension schemes around the world in the 20th century substantially reduced old-age poverty and inequality. Part of the fall in poverty rates occurred because money was transferred from high-income to low-income people, which is the fundamental philosophy behind welfare-based pension schemes.
However, part occurred because people were forced to pay taxes or make compulsory savings when they were young and middle aged, and were provided with a pension in turn when they were old. These transfers are basically from a younger person to their older self, rather than from one person to another. They are a major component of the decrease in old-age poverty in countries that have contributory pension schemes.
Because government pension schemes provide retirement income until a person dies, they also help reduce poverty by dealing with several types of risk and uncertainty. The first risk is a nice problem to have – that your money runs out because you live “too long”.
The second risk is that a person’s private investment returns are poor, leaving them with too little money in retirement. In the 20th century inflation was a major cause of this type of risk.
A third risk is that a person’s family arrangements change in ways that significantly raise expenses. A person could get divorced for example, or they or their spouse could have very large health expenses.
A fourth risk, for those people who plan to work in old age to support themselves, is that bad health or poor “late-life earning opportunities” leave them without enough income.
Governments are very good at managing some of the investment and longevity risks associated with retirement because they can share risks across generations in ways that are not possible for individuals, families or even big insurance companies. Governments do not die or go bankrupt. Moreover, they have the legal powers to make and enforce rules so they can ensure old people get some money if disasters occur.
If things go wrong, governments can transfer money to older people by raising taxes on others. If things go really wrong, they can borrow and push part of the bill on to other generations.
Pay-as-you-go schemes are particularly good at shifting resources across generations – too good, really, as was discussed last week - but save-as-you-go schemes can also be designed to allow different cohorts to share risk. These advantages are offset by a different risk, which is probably the most difficult to manage: that the government changes the rules of the pension scheme, leaving people with a much smaller retirement income than they anticipated.
New Zealand is unique in the OECD because the government pension scheme is only based on welfare principles, not contributory principles. To many people, one of the most attractive aspects of the New Zealand Superannuation scheme is that everyone gets the same amount, independent of their current or previous incomes or the amount they have saved. No matter someone’s health, their age, gender, ethnicity, their current income or their lifetime income, everyone meeting residency requirements is entitled to the same payment.
It also provides a measure of dignity, for entitlement is based on age and nationality, not any arbitrarily defined set of achievements. In many countries, universality is a highly prized alternative to what can otherwise be demeaning and intrusive criteria to obtain a small welfare benefit (see for example Diamond and Mirrlees 1978; Kildal and Kuhnle 2008.
Most countries have a small “welfare” based pension for people whose “contributory” retirement income would otherwise be inadequate. Older generations of New Zealanders chose not to have a contributory scheme but have a welfare pension that is relatively generous by the standards of most OECD countries.
This has advantages and disadvantages. One of the advantages is that it helps to substantially reduce old-age poverty. However, it exacerbates other dimensions of inequality. New Zealand Superannuation keeps older people out of poverty because it is set at a relatively high level relative to average wages.
This means a lot of tax revenue is needed, and in New Zealand this is raised from general funds not specific social security taxes. As I shall discuss over the next few articles, these taxes not only reduce the incomes of younger people, but they also change the prices of all sorts of goods and services – including housing.
By reducing income and raising house prices, New Zealand’s tax and pension system delays home ownership and raises rents. This can increase inequality elsewhere in society even as it reduces inequality amongst older people.
New Zealand Superannuation may also be increasing long- run wealth inequality. This argument is relatively new, and was developed by four of the world’s leading retirement income experts Jagadeesh Gokhale, Laurence Kotlikoff, James Sefton and Martin Weale. It concerns the ways public pension schemes and private savings counteract the risk of dying relatively young.
Long run wealth inequality
Under the current pension system, many people pay taxes for their entire lives but then receive little or no pension benefits because they die young. To some extent, of course, this is a feature and not a bug: the pension is provided as an annuity precisely so that people know that they will have a retirement income as long as they live.
Inevitably this means that people who draw the short straw in life and die young receive fewer total benefits than those who live a long time. This can be problematic, however, when some groups of people systematically die younger than average for then the system seems rigged against them. If people who die young also have lower incomes, it can feel even worse. They can complain that they have paid taxes on low incomes all their working lives, even though on average they won’t live long enough to enjoy a full pension.
Are there groups of people who on average have low incomes and die young? Yes. For example, in New Zealand Māori and Pacific people have lower life expectancy than Pakeha or Asian people, and also have lower average earnings. But it is much more widespread than this. In most developed countries, if you look at people born in a particular year and consider men and women separately, people who have low incomes typically have lower life expectancy than those who have high incomes.
The most comprehensive study examining this issue was conducted in the USA using tax records. On average, a woman aged 40 at the 20% position in the female income distribution lives an additional 43 years, versus 47 years for a woman at the 80% position. For a man aged 40 the respective figures are 38 years and 44 years (See Chetty et al (2016).
This relationship between life expectancy and income has major implications for the way a pay-as-you-go pension scheme such as New Zealand Superannuation affects long-term wealth distribution.
This, of course, is not the full story. Women live several years longer than men on average, but earn less over their lifetimes. These differences makes it difficult to answer whether it is fair that groups of people with high life expectancy are taxed in the same way as people with low-life expectancy but get the pension for a longer length of time. If you compare men and women, resources on average are transferred from high-income, low- life expectancy men to low-income high-life expectancy women. (I don’t know if this is fair, and I am not offering an opinion!)
However if you compare men and women separately, the lower taxes paid by low income pension are offset by the lower total pension benefits they receive.
Gokhale and his coauthors realized that when people with low incomes have low life expectancy, tax-funded pension schemes may raise long-run wealth inequality even though they reduce short run income inequality. This happens because a person’s entitlement to the pension is not normally extended to their spouse or children if they die young.
Relatively few people who had low incomes during their working lives have many assets other than their house and their entitlement to a government pension when they are old. In contrast, high income people have a house, their entitlement to a government pension, and a large quantity of other bequeathable assets such as investment properties, shares, or bank accounts.
Consequently, high income people have a much greater fraction of their total retirement assets (including their government pension entitlement) in a bequeathable form than low-income people.
Gokhale, Kotlikoff, Sefton and Weale carefully simulated how the positive correlation between income and longevity affects the long run distribution of wealth. Taking into account the number of children different couples have, and the increased probability that low-income people die young, their simulations suggest that tax-financed pension schemes worsen long-run wealth inequality by reducing the amount of bequeathable assets that low-income people have if they die young. This reduction in bequeathable assets sets up a cycle of low wealth and intergenerational inequality.
As noted above, Māori and Pacific people have lower life expectancy that Pakeha or Asian people, and also have lower average earnings. This suggests that if it were possible to redesign New Zealand Superannuation to increase the amount of bequeathable wealth in the event of an earlier death, some of the main beneficiaries would be the families and descendants of Māori and Pacific people.
It is possible to sidestep this bequeathable wealth trap by financing a larger portion of retirement income through a compulsory saving scheme. The trick is to fund a “base-load” fraction of the government pension from a personal compulsory scheme, while funding the rest from government taxes.
For example, the personal compulsory saving scheme may be used to provide the first 40% of someone’s retirement income, and the government provides the rest. The base-load component would be bequeathable.
In this case it is still bad luck to die young, but the deceased person’s partner or children will inherit any unspent portion of the funds in the compulsory saving scheme. Since many people have a strong desire to leave assets to their family when they die, this is an attractive component of a compulsory saving scheme.
Towards the end of the series I will suggest a method of changing New Zealand’s retirement saving scheme so it increases the bequeathable portion of a person’s retirement assets without reducing the minimum retirement income they receive each year. This represents a way of improving the current system that may be appreciated both by those who systematically die younger than others (a group that is disproportionately male, Māori and Pacific) and those who could benefit from a larger estate if their spouses or parents die young.
In other words, there is a way of keeping most of the advantages of New Zealand Superannuation while reducing one of its disadvantages. This prospect sounds appealing and is one of the reasons why a rethink of New Zealand’s retirement arrangements is in order.
It is not the only reason. New Zealand’s retirement income arrangements are also increasing inequality in at least two other ways. These are tackled next week.
*This series and an accompanying paper are based on work I started in 2020 with Jeanne-Marie Bonnet while we were both at the University of Otago. I am very grateful for her assistance and insights. All errors remain my own.
(This article is part 6 in the series. Part 1 is here, part 2 is here, part 3 is here, part 4 is here, and part 5 is here).
**Andrew Coleman is a visiting professor at the Asia School of Business. This article is his personal view of retirement policy in New Zealand, based on academic study.
Coleman is on extended leave from the Reserve Bank of New Zealand, while working overseas. The views expressed in this article do not represent the RBNZ and are unrelated to work conducted at the Bank, which has no responsibility for retirement policy in New Zealand.
We welcome your comments below. If you are not already registered, please register to comment
Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.